Analysis of large antenna systems

The research presented in this thesis has been conducted within the framework of the Square Kilometre Array (SKA) project. SKA is a next generation radio telescope that will have a receiver sensitivity two orders of magnitude larger than the most sensitive radio telescope currently in operation. To meet the specifications, various low-cost low-noise actively beamformed receiving array antennas are being considered. A major problem in designing these systems is that the present-day commercially available electromagnetic solvers need an excessive amount of memory and simulation time to solve electrically large antenna problems. Moreover, it is essential to be able to analyze the receiver sensitivity of large antenna array systems to understand the sensitivity limiting factors. No dedicated commercial software tools exist that can analyze the receiver sensitivity of entire antenna systems specifically for radio astronomy. The thesis subject deals with two major challenges: (i) To accurately compute the impedance and radiation characteristics of realistically large and complex antenna arrays using only moderate computing power, particularly, of single and dual-polarized arrays of 100+ Tapered Slot Antenna (TSA) elements that are electrically interconnected. If the collection of these elements forms a subarray of a larger system, it is also of interest to analyze an array of disjoint subarrays. (ii) To characterize the system sensitivity of actively beamformed arrays of strongly coupled antenna elements. To address the above challenges, a conventional method-of-moments approach to solving an electric-field integral equation is enhanced using the Characteristic Basis Function Method (CBFM) to handle electrically large antenna problems. The generation of the associated reduced matrix equation is expedited by combining the CBFM with the Adaptive Cross Approximation (ACA) technique. Furthermore, because an overlapping domain decom270 Bibliography position technique is employed, Characteristic Basis Functions (CBFs) are generated that partially overlap to ensure the continuity of the current between adjacent subdomains that are electrically interconnected. While generating the CBFs, edge-singular currents are avoided by a post-windowing technique. Finally, a meshing strategy is proposed to optimally exploit the quasi-Toeplitz symmetry of the reduced moment matrix. The numerical accuracy and efficiency has been determined for numerous cases, among which a dual-polarized interconnected TSA array of 112 elements that has been fabricated and subsequently validated by measurements. The receiver system has been modeled by both a numerical and a semi-analytical method. The models account for a nonuniform brightness temperature distribution of the sky, mismatch effects, noise that emanates from amplifiers inputs and re-enters the system coherently through the mutually coupled antennas (noise coupling), beamformer weights, etc. Results are shown for a practical setup and design rules are derived which demonstrate that minimum receiver noise can be reached by noise matching the low-noise amplifiers to the active antenna reflection coefficient, rather than the passive one. Finally, it is demonstrated that the radiation efficiency of antennas is an important quantity that can degrade the system sensitivity severely. Nevertheless, a number of commercial software tools have shown to be inadequate as the computed efficiency exceeds 100%. A method is proposed which is numerically efficient and robust since it guarantees an efficiency below 100%.